
*use "anes 04.dta"
*note: no evangelical variable

regress sorting religiosity01 pid3 pid3Xrelig whiteevangelical secular mainline jewish catholic interest   abortion govspend sophistication male white black age educ7cat oldsouth [pweight= vcf0010z]
